The effect of expectations on the Brazilian Central Bank’s policy rate

Abstract

This article investigates how expectations influence the determination of the Brazilian benchmark interest rate based on Keynes’ views on the relationship between expectations and monetary policy. Then, as empirical methodology, Autoregressive Distributed Lag Models and Bounds Testing Approach to Cointegration are used to study, in the short and long-run, the connection between expectations and central bank interest rate. For the period from 2001Q3 to 2017Q4, results show that in the long-run business and consumer confidence, as well as expectations related to market interest rates, GDP, inflation and exchange rate play an important role on monetary policy actions. In the short-run, business and consumer confidence lose importance, while the other mentioned expectations are still statistically significant. The findings of the paper lend credence to Keynes’s view on the relation between expectation and money policy in Brazil.

Keywords:

• Brazilian economy
• expectations
• interest rates
• monetary policy

Notes

1 The Selic rate is used as a reference for daily financing rates secured by financial institutions using Brazilian government bonds and calculated by the Special System for Settlement and Custody (Selic).

2 It is worth mentioning the fourth reason Keynes (1937) gave to explain money demand, the finance motive. Entrepreneurs request liquidity for this reason when they plan an investment and need funding for that. Self-financing partially addresses this need, but the remaining part needs to be financed by the financial system. The finance motive is part of the overall money demand, so it plays a role on determining market interest rate, as firms’ demand for financing faces the liquidity pool offered by financial markets.

3 To our model, it is not necessary to emphasize the special role of banks as money creators in monetary economies, still it is not adequate to depict their behavior through relations (3), (3.1) and (4), as their $L\left(r_m\right)$ is not inactive balance, only $M_p^D$ can be called so. So, the following relation for banks’ precautionary and speculative demand for money can be written: $M_s^S=\frac{1}{M_p^D}.$ As mentioned before, $M_s^S$ must be understood as money supply for those seeking liquidity in financial markets. Arguments for $M_s^S$ and $M_p^D$remain the same and this relation means that, when banks have greater liquidity preference, they reduce their speculative actions, curtailing money supply and pushing interest rates up.

4 We present only a simple model that displays how an agent sets an interest rate on some asset. This model can be expanded to all agents with their desired interest rates. Let $yc$ be the yield curve, i an agent and j a type of asset. We derive the financial system yield curve ($yc)$ as $yc=\sum _{ij}{(1+\phi ){\Delta r}_{ij}}^{\epsilon }+r_{\mathit{cbt}+1ij}^{\epsilon }.$ Moreover, we are assuming the central bank base rate as the only benchmark rate of the financial system. Although it is not the only benchmark rate, it is the most relevant one. Even interest rates on fiscal policy bonds, which form a yardstick for long-term private debt negotiations, set a spread over the base rate.

Additional information

Funding

This author thanks CNPq, CAPES and FAPEMIG for the financial support to this research.

Notes on contributors

Fábio Henrique Bittes Terra

Fábio Henrique Bittes Terra is at Federal University of ABC, Sao Bernardo do Campo, Brazil; Federal University of Uberlandia, Uberlandia, Brazil; CNPq Researcher, Brasilia, Brazil.

Cleomar Gomes

Cleomar Gomes is at Federal University of Uberlandia, Uberlandia, Brazil; CNPq Researcher, Brasilia, Brazil.